Status: Ongoing analysis Material from: Linus, Mohamed
1. Impact
In March 2021 eastern Australia was hit by extreme rainfall.
https://www.bbc.co.uk/news/world-australia-56476998
The news severe flooding was due to sustained heavy rainfall that started to affect the area around the 17th of March over an already wet catchments. The heavy precipitations were associated with a slow moving trough with a sustained feed of tropical moisture from the North West. In the reports received at ECMWF the maximum daily totals exceeded 350mm but the accumulations over 5 days exceeds 1000 mm in few places and 500mm in many locations. ECMWF forecasts (from HRES and ENS) managed to predict the evolution of the event many days in advance.
2. Description of the event
The plots below show analyses of MSLP and 6-hour forecasts of precipitation. The first plot is from 18 March 00UTC and last 21 March 12UTC, every 12 hours.
The plots below show the precipitation observations accumulated on 18, 19, 20, 21 March and summary statistics.
The plot below shows timeseries of rainfall stats (Mean, Max and number of station reporting more than 10mm) in a box in New South Wales.
3. Predictability
3.1 Data assimilation
3.2 HRES
The plots below show HRES forecasts of 3-day precipitation 19-21 March.
3.3 ENS
The plots below show EFI for 3-day precipitation for forecasts valid 19-21 March.
The plot below shows the forecast evolution for 3-day precipitation 19-21 March averaged inside the box around Sidney outlined in the plots above. The plot includes HRES (red dot), ENS control (purple dot), ENS distribution (blue boxes, median square) and model climate distribution (red boxes).
3.4 Monthly forecasts
The plots below show the precipitation anomaly for forecasts valid 15-21 March.
3.5 Comparison with other centres
3.6 Glofas
The plots below show Glofas forecasts of probability for 20-year return period exceedance in the forecasts from 16, 18, 19 and 20 March.
The plot below shows the hydrography in the forecast from 21 March for a river near Sidney.
4. Experience from general performance/other cases
5. Good and bad aspects of the forecasts for the event
- Although the predicted amounts were large but there were significantly smaller than the observed range.
- The Severity of the event started to be seen in the forecast only few days in advance. The EFI and other products were after that clearly showing the correct pattern of this extreme events.
- The large scale wet anomaly was correctly identified by the extended range forecasts only after the 4th of March. The forecasts issued on the 8th of March significantly reduced the signal with a northward shift
- Glofas managed to predict the risk of floods for catchments in the areas but also far in land. Being fed by the ENS rainfall Glofas did provide warning only few days in advance